The purpose is to compare the contribution of gluconeogenesis to hepatic glucose production of preterm infants given glucose alone vs. a combination of either glucose and amino acids, glucose and lipid or glucose, amino acids and lipid during the first 5 days of life. The hypotheses to be tested are: amino acids, lipid and a combination of both enhance gluconeogenesis in the VLBW infant; substitution of IV amino acids, lipid or both for glucose diminish hyperglycemia without increasing the risk of hypoglycemia. (NRSU protocol) In a first part of the project gluconeogenesis was estimated in three comparable groups of extremely premature infants receiving parenteral nurtition, employing three independent methods using [U13C]glucose (n=8), [2-13C]glycerol (n=6), or [2H2O] (n=6). Twenty infants gestational age 27 +/- 0.2 weeks (mean+/-SE), birth weight 996 +/- 28 grams, received a 8-12hrs parenteral infusion of lipid and amino acids, while the glucose infusion rate was reduced to about half the glucose turnover rate to revent hyperglycemia. The results demonstrate that the infants remained normoglycemic throughout the study periods. The [U-13C]glucose and the [2-13C]glycerol tracers provided equal estimates of gluconeogenesis averaging 72.3 +/- 4.5 and 72.4 +/- 9.1% of glucose production. Glycerol was the principal gluconeogenic substrate. The [2H2O] estimate, which does not include the contribution from glycerol, was comparable to the non-glycerol contribution measured by the [2-13C]glycerol method. Thus, after 4 hrs of adequate supply of lipids and amino acids but reduced glucose delivery, normoglycemia was maintained, gluconeogenesis was the principle source of glucose production and glycerol the primary gluconeogenic substrate. (Manuscript in progress). In the second part of the project, the impact of parenterally administrated lipids and amino acids on gluconeogenesis will be tested separately using [2-13C]glycerol and [U-13C]glucose, respectively. Preliminary results obtained in a group of extremely premature infants comparable to those included in the first part of the project and, receiving the same parenteral nutrition except the amino acids were removed, indicate that removal of amino acids does not impair gluconeogenesis. These studies are ongoing.